We published estimates of the risk of breast cancer and of ovarian cancer from a study of 5,300 Jewish men and women in the Washington D.C. area, using a novel approach, the kin- cohort design, that we developed. In the study, we collected blood to determine whether the volunteer was a carrier of one of two BRCA1 and one BRCA2 mutations. We also obtained a detailed family history of breast and ovarian cancer in the mother, sisters, and daughters of the volunteers. By using principles of inheritance, our method was able to develop age-specific estimates of cancer risk in mutation carriers. We also developed the analytic techniques for this design and have submitted a paper describing its strengths and limitations. We participated in a study linking benzene poisoning, a risk factor for hematologic malignancy, with the NQO1 609C->T mutation and rapid fractional excretion of chlorzoxazone. This is the first reported evidence of genetic susceptibility in humans to a disease associated with benzene. To determine the feasibility of studying the role of endogenous hormones for a variety of reproductive cancers, we conducted assay reliability studies that determine which laboratories perform well and define the precision of the assays. Assays for serum estrone, estradiol and estrone sulfate are sufficiently precise to permit reliable comparisons among women in a given menstrual phase. Related reliability studies were completed for estrogen metabolism in urine. We also worked on studies of reliability of laboratory assays for human papillomavirus.